Support assignment and arithmetic operations (+=)

[mihaibudiu]

Personnel

• Owner: @ChrisDodd
• Supporters: @mbudiu-vmw @jafingerhut

Design

• Document:

Implementation

• p4-spec: Old PR that needs updating: Add definition for compound assignment operators #1144
• p4c: PR Support for op= assignments in the frontend/midend p4c#5108 (older PR, probably abandoned effort: Add support for compound assignment operators p4c#3669 - incomplete as of 2023-Jun-05)

Process

• LDWG discussed:
• LDWG approved:
• Merged into p4-spec:
• Merged into p4c:

=======================================

[apinski-cavium]

I am trying to remember if we have a case where a lvalue expression can have a side effect which will need to be handled correctly in the specifications and the compilers. That is my only concern here.
An example where a side effect could happen I suspect is in header stack indexing:

hdr_stack[func(h)].field += 4;

But that could be resolved by doing:

bit tmp1 = func(h);
hdr_stack[tmp1].field = hdr_stack[tmp1].field + 4;

Candidates for expressions to do similarly to +=:

-=
&=
|=
^=
*=
/=
<<=
>>=

[blp]

C specifies that a += b is equivalent to a = a + b except that a is evaluated only once. C99 says, for example:

A compound assignment of the form E1 op = E2 differs from the simple assignment expression E1 = E1 op (E2) only in that the lvalue E1 is evaluated only once.

It would make sense to me to specify compound assignment in P4 in a similar way.

[rst0git]

@mbudiu-vmw, @apinski-cavium and @blp thank you for your input! I have opened the following pull request to specify compound assignment operators in the P4_16 spec: #1144

[jnfoster]

In the interest of tidying up the set of active issues on the P4 specification repository, I'm marking this as "stalled" and closing it. Of course, we can always re-open it in the future if there is interest in resurrecting it.

[jafingerhut]

Reopening this issue since there is new interest in implementing it in p4c: p4lang/p4c#5108

[vlstill]

PR: #1359